Explore the vast range of titles available.

Artificial neural networks (ANNs) inspired by biology are beginning to be widely used to model behavioural and neural data, an approach we call ‘neuroconnectionism’. ANNs have been not only lauded as the current best models of information processing in the brain but also criticized for failing to account for basic cognitive functions. In this Perspective article, we propose that arguing about the successes and failures of a restricted set of current ANNs is the wrong approach to assess the promise of neuroconnectionism for brain science. Instead, we take inspiration from the philosophy of science, and in particular from Lakatos, who showed that the core of a scientific research programme is often not directly falsifiable but should be assessed by its capacity to generate novel insights. Following this view, we present neuroconnectionism as a general research programme centred around ANNs as a computational language for expressing falsifiable theories about brain computation. We describe the core of the programme, the underlying computational framework and its tools for testing specific neuroscientific hypotheses and deriving novel understanding. Taking a longitudinal view, we review past and present neuroconnectionist projects and their responses to challenges and argue that the research programme is highly progressive, generating new and otherwise unreachable insights into the workings of the brain.Artificial neural networks are being widely used to model behavioural and neural data. In this Perspective article, Doerig et al. present neuroconnectionism as a Lakatosian research programme using artificial neural networks as a computational language for expressing falsifiable theories and hypotheses about the brain computations underlying cognition.

Recently, deep neural networks (DNNs) have been shown to provide accurate predictions of neural responses across the ventral visual pathway. We here explore whether they also provide accurate predictions of neural responses across the dorsal visual pathway, which is thought to be devoted to motion processing and action recognition. This is achieved by training deep neural networks to recognize actions in videos and subsequently using them to predict neural responses while subjects are watching natural movies. Moreover, we explore whether dorsal stream representations are shared between subjects. In order to address this question, we examine if individual subject predictions can be made in a common representational space estimated via hyperalignment. Results show that a DNN trained for action recognition can be used to accurately predict how dorsal stream responds to natural movies, revealing a correspondence in representations of DNN layers and dorsal stream areas. It is also demonstrated that models operating in a common representational space can generalize to responses of multiple or even unseen individual subjects to novel spatio-temporal stimuli in both encoding and decoding settings, suggesting that a common representational space underlies dorsal stream responses across multiple subjects. •Deep neural network and dorsal stream representations show correspondence.•Dorsal stream representations are shared between subjects.•A common encoder can predict fMRI responses to novel stimuli for unseen subjects.•A common decoder can identify novel stimuli from fMRI responses for unseen subjects.

It is well established that neuronal oscillations at different frequencies interact with each other in terms of cross-frequency coupling (CFC). In particular, the phase of slower oscillations modulates the power of faster oscillations. This is referred to as phase–amplitude coupling (PAC). Examples are alpha phase to gamma power coupling as observed in humans and theta phase to gamma power coupling as observed in the rat hippocampus. We here ask if the interaction between alpha and gamma oscillations is in the direction of the phase of slower oscillations driving the power of faster oscillations or conversely from the power of faster oscillations driving the phase of slower oscillations. To answer this question, we introduce a new measure to estimate the cross-frequency directionality (CFD). This measure is based on the phase-slope index (PSI) between the phase of slower oscillations and the power envelope of faster oscillations. Further, we propose a randomization framework for statistically evaluating the coupling measures when controlling for multiple comparisons over the investigated frequency ranges. The method was firstly validated on simulated data and next applied to resting state electrocorticography (ECoG) data. These results demonstrate that the method works reliably. In particular, we found that the power envelope of gamma oscillations drives the phase of slower oscillations in the alpha band. This surprising finding is not easily reconcilable with theories suggesting that feedback controlled alpha oscillations modulate feedforward processing reflected in the gamma band. •A new measure to estimate cross-frequency directionality is introduced.•A randomization framework for statistically evaluating the coupling measures is proposed.•The power envelope of gamma oscillations is shown to drive slower oscillations in the alpha band.

The aim of this study was to explore modifications of functional connectivity in multiple resting-state networks (RSNs) after moderate to severe traumatic brain injury (TBI) and evaluate the relationship between functional connectivity patterns and cognitive abnormalities. Forty-three moderate/severe TBI patients and 34 healthy controls (HC) underwent resting-state fMRI. Group ICA was applied to identify RSNs. Between-subject analysis was performed using dual regression. Multiple linear regressions were used to investigate the relationship between abnormal connectivity strength and neuropsychological outcome. Forty (93%) TBI patients showed moderate disability, while 2 (5%) and 1 (2%) upper severe disability and low good recovery, respectively. TBI patients performed worse than HC on the domains attention and language. We found increased connectivity in sensorimotor, visual, default mode (DMN), executive, and cerebellar RSNs after TBI. We demonstrated an effect of connectivity in the sensorimotor RSN on attention ( p  < 10 −3 ) and a trend towards a significant effect of the DMN connectivity on attention ( p  = 0.058). A group-by-network interaction on attention was found in the sensorimotor network ( p  = 0.002). In TBI, attention was positively related to abnormal connectivity within the sensorimotor RSN, while in HC this relation was negative. Our results show altered patterns of functional connectivity after TBI. Attention impairments in TBI were associated with increased connectivity in the sensorimotor network. Further research is needed to test whether attention in TBI patients is directly affected by changes in functional connectivity in the sensorimotor network or whether the effect is actually driven by changes in the DMN.

Recent experiments have revealed a hierarchy of time scales in the visual cortex, where different stages of the visual system process information at different time scales. Recurrent neural networks are ideal models to gain insight in how information is processed by such a hierarchy of time scales and have become widely used to model temporal dynamics both in machine learning and computational neuroscience. However, in the derivation of such models as discrete time approximations of the firing rate of a population of neurons, the time constants of the neuronal process are generally ignored. Learning these time constants could inform us about the time scales underlying temporal processes in the brain and enhance the expressive capacity of the network. To investigate the potential of adaptive time constants, we compare the standard approximations to a more lenient one that accounts for the time scales at which processes unfold. We show that such a model performs better on predicting simulated neural data and allows recovery of the time scales at which the underlying processes unfold. A hierarchy of time scales emerges when adapting to data with multiple underlying time scales, underscoring the importance of such a hierarchy in processing complex temporal information.

Evaluation of the structural connectivity (SC) of the brain based on tractography has mainly focused on the choice of diffusion model, tractography algorithm, and their respective parameter settings. Here, we systematically validate SC derived from a post mortem monkey brain, while varying key acquisition parameters such as the b-value, gradient angular resolution and image resolution. As gold standard we use the connectivity matrix obtained invasively with histological tracers by Markov et al. (2014). As performance metric, we use cross entropy as a measure that enables comparison of the relative tracer labeled neuron counts to the streamline counts from tractography. We find that high angular resolution and high signal-to-noise ratio are important to estimate SC, and that SC derived from low image resolution (1.03 mm3) are in better agreement with the tracer network, than those derived from high image resolution (0.53 mm3) or at an even lower image resolution (2.03 mm3). In contradiction, sensitivity and specificity analyses suggest that if the angular resolution is sufficient, the balanced compromise in which sensitivity and specificity are identical remains 60–64% regardless of the other scanning parameters. Interestingly, the tracer graph is assumed to be the gold standard but by thresholding, the balanced compromise increases to 70–75%. Hence, by using performance metrics based on binarized tracer graphs, one risks losing important information, changing the performance of SC graphs derived by tractography and their dependence of different scanning parameters.

Eye movements are an integral part of human perception, but can induce artifacts in many magneto-encephalography (MEG) and electroencephalography (EEG) studies. For this reason, investigators try to minimize eye movements and remove these artifacts from their data using different techniques. When these artifacts are not purely random, but consistent regarding certain stimuli or conditions, the possibility arises that eye movements are actually inducing effects in the MEG signal. It remains unclear how much of an influence eye movements can have on observed effects in MEG, since most MEG studies lack a control analysis to verify whether an effect found in the MEG signal is induced by eye movements. Here, we find that we can decode stimulus location from eye movements in two different stages of a working memory match-to-sample task that encompass different areas of research typically done with MEG. This means that the observed MEG effect might be (partly) due to eye movements instead of any true neural correlate. We suggest how to check for eye movement effects in the data and make suggestions on how to minimize eye movement artifacts from occurring in the first place.

Quasi-experimental research designs, such as regression discontinuity and interrupted time series, allow for causal inference in the absence of a randomized controlled trial, at the cost of additional assumptions. In this paper, we provide a framework for discontinuity-based designs using Bayesian model averaging and Gaussian process regression, which we refer to as ‘Bayesian nonparametric discontinuity design’, or BNDD for short. BNDD addresses the two major shortcomings in most implementations of such designs: overconfidence due to implicit conditioning on the alleged effect, and model misspecification due to reliance on overly simplistic regression models. With the appropriate Gaussian process covariance function, our approach can detect discontinuities of any order, and in spectral features. We demonstrate the usage of BNDD in simulations, and apply the framework to determine the effect of running for political positions on longevity, of the effect of an alleged historical phantom border in the Netherlands on Dutch voting behaviour, and of Kundalini Yoga meditation on heart rate.

After the presentation of a visual stimulus, neural processing cascades from low-level sensory areas to increasingly abstract representations in higher-level areas. It is often hypothesised that a reversal in neural processing underlies the generation of mental images as abstract representations are used to construct sensory representations in the absence of sensory input. According to predictive processing theories, such reversed processing also plays a central role in later stages of perception. Direct experimental evidence of reversals in neural information flow has been missing. Here, we used a combination of machine learning and magnetoencephalography to characterise neural dynamics in humans. We provide direct evidence for a reversal of the perceptual feed-forward cascade during imagery and show that, during perception, such reversals alternate with feed-forward processing in an 11 Hz oscillatory pattern. Together, these results show how common feedback processes support both veridical perception and mental imagery.

Encoding models are used for predicting brain activity in response to sensory stimuli with the objective of elucidating how sensory information is represented in the brain. Encoding models typically comprise a nonlinear transformation of stimuli to features (feature model) and a linear convolution of features to responses (response model). While there has been extensive work on developing better feature models, the work on developing better response models has been rather limited. Here, we investigate the extent to which recurrent neural network models can use their internal memories for nonlinear processing of arbitrary feature sequences to predict feature-evoked response sequences as measured by functional magnetic resonance imaging. We show that the proposed recurrent neural network models can significantly outperform established response models by accurately estimating long-term dependencies that drive hemodynamic responses. The results open a new window into modeling the dynamics of brain activity in response to sensory stimuli.